implement
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process external, potentially untrusted data that could influence the agent's behavior through embedded instructions in tickets or specifications.
- Ingestion points: The agent is instructed to read requested specs or tickets as the primary source of truth for implementation tasks (SKILL.md).
- Boundary markers: There are no explicit instructions to treat ticket/spec content as data rather than instructions, nor are there delimiters defined to separate user instructions from ticket content.
- Capability inventory: The skill has significant write-access capabilities, including implementing code slices via TDD, performing domain modeling changes, committing results to the repository, and opening Pull Requests (SKILL.md).
- Sanitization: The instructions do not define any sanitization, filtering, or validation steps for the content of the specifications before implementation begins.
Audit Metadata